Allocation biopsies of deceased donor kidneys: a necessary tool to expand the donor pool

The increasing incidence and prevalence of kidney failure globally has corresponded to a marked increase in the demand for kidney transplantation. However, because the number of donated organs is insufficient to meet this demand, most patients with kidney failure never experience the large survival and quality of life benefits that transplantation confers. Expanding access to transplantation requires an expansion of the deceased donor organ pool, including the procurement and utilization of organs previously considered unacceptable, such as those from donors with diabetes or dialysis-dependent acute kidney injury. Nevertheless, many potentially transplantable kidneys are not recovered or transplanted, as best evidenced by a nearly 30% rate of discard of recovered kidneys in the United States.¹ Combating this underutilization depends on improving clinicians’ confidence in the viability of transplantable organs that have undesirable characteristics.

Organ quality assessments often aim to augment donor clinical data with histologic information from a biopsy performed during the allocation process, referred to interchangeably as a “procurement,” “allocation,” or “preimplantation” biopsy. Although histologic information that includes details on glomerulosclerosis, interstitial fibrosis, tubular atrophy, arteriosclerosis, and arteriolosclerosis with hyalinosis can provide additional insights, there are inherent challenges to incorporating this information into a clinical decision.² Accurate and reliable histologic assessments require significant expertise and, even then, are subject to challenges of reproducibility and the interpretation of test results with varying pretest probabilities.^3,4 As a result, although histology can provide additional insights about the eventual function of a transplanted allograft, this information is most useful in circumstances where additional data would help “rule in” an organ that would otherwise be discarded based on clinical information (or lack thereof) alone.

Preimplantation histology can provide additional prognostic information beyond donor clinical data

The increasing prevalence of chronic conditions, such as diabetes, hypertension, and obesity, in the general population and among deceased donors adds complexity to the evaluation of donor organ quality. However, clinician gestalt of organ quality provides almost no additional prognostic information above the use of clinical data alone.⁵ The use of accurately assessed histology is, therefore, a valuable supplement for kidneys from donors with features that would otherwise favor discard, such as severe acute kidney injury, or uncertain duration of diabetes/hypertension. This principle extends from native kidney biopsies, where more severe chronic parenchymal scarring is associated with lower kidney function and higher rate of adverse kidney outcomes,⁶ and “chronicity scores” often inform therapeutic decisions by providing an ascertainment of recoverable residual kidney mass.

Given the broadly accepted prognostic value of chronic changes in patients with kidney disease, it would seem unlikely that histology during allocation, if reported accurately, is inherently different. Several retrospective studies, however, have suggested the absence of a relationship between preimplantation biopsy results and post-transplant outcomes after accounting for donor clinical characteristics. Although some of these findings result from misclassification due to inaccurate or incomplete reporting, there are likely other confounders inherent to the design of these analyses.³ Because retrospective studies are restricted to organs that were both biopsied and transplanted, they tend not to include donors with both suboptimal histology and unfavorable clinical characteristics, which would likely have concordant poor post-transplant outcomes—thereby biasing these retrospective cohorts toward a null effect. Unfortunately, selection bias impacts even cohorts from outside the United States, where biopsy information is less often part of the allocation process. For example, a recent study demonstrating a lack of association between biopsy findings and post-transplant outcomes in France and Belgium included few marginal organs, thus limiting its implications for marginal donors, where histologic assessment would have most value: for example, among the Belgian validation cohort, median donor age was only 48 years and <1% had diabetes.⁷ A more recent analysis of US national data showed that limiting analyses to kidneys that are likely to be both biopsied and transplanted demonstrates a clear independent association between more advanced chronic changes and a higher rate of allograft failure.⁸

Furthermore, although there is evidence that allocation-biopsy–specific techniques that oversample subcapsular tissue or rely on non-kidney pathologists contribute to poor biopsy reproducibility, there is no reason that biopsies must be performed or interpreted in these manners. In fact, among kidneys where initial biopsy findings were not associated with outcomes, repeating biopsies using standardized technique or rereading initial slides using a pathologist with kidney expertise yields findings with significant prognostic value.^3,4 In other words, the inferior prognostic value of biopsy results obtained using suboptimal sampling, processing, or interpretation techniques should not be inferred to apply to higher-quality allocation biopsies, and the variation in clinical practice between organ procurement organizations only further confounds the ability to interpret retrospective cohorts when this is not considered.

Allocation biopsies improve the utilization of marginal quality organs

The most frequent (and valid) criticism of allocation biopsies is their use in circumstances where additional histologic information is not needed and can only confound decision-making and have a detrimental impact on the utilization of high-quality organs. Such organs do not need an allocation biopsy. Rather, allocation biopsies are most valuable as an additional source of objective data in situations where clinical data are limited (e.g., a donor with high creatinine with unknown medical history) or conflicting.^2,9 There is no high-quality evidence that demonstrates that simply performing biopsies on recovered kidneys increases the risk of organ discard. Prior analyses that have suggested an association between performing a biopsy and increased the risk of discard overlook the fact that retrospective studies include few high-quality kidneys, leading to significant bias by virtue of selection in which kidneys are being biopsied.¹⁰ In fact, the same national analysis demonstrated that discard rates were similar for Organ Procurement Organizations (OPOs) that had the highest and lowest proportion of kidneys that underwent a biopsy.¹⁰

Confirmation of the value of biopsies in “rescuing” marginal organs from discard comes from 2 simulated organ offer studies demonstrating that the absence of biopsy results does not improve transplant centers’ willingness to accept marginal organs. A survey of Canadian transplant physicians showed similar low acceptance rates for a marginal donor kidney (hypertension, diabetes, smoking history, and urine albumin-to-creatinine ratio 100 mg/g) when no biopsy was presented and when a biopsy showing 25% glomerulosclerosis without arterial hyalinosis was presented.¹¹ Similarly, in a US simulated organ offer study, offer acceptance rates for donors with elevated creatinine were low when no biopsy was presented but significantly improved when favorable biopsy results were presented.¹² These data are consistent with a US survey in which 68% of transplant centers agreed with the statement that “A ‘good’ biopsy will often persuade me to transplant a kidney I would consider marginal based on clinical characteristics alone.”⁹

Histology-informed allocation improves outcomes of transplant using marginal quality organs

There are few prospective data on the use of allocation biopsies to inform utilization decisions, although 2 available studies support histology-informed allocation of organs from marginal donors. In a prospective Italian trial, histology-informed kidney allocation resulted in achieving post-transplant outcomes using kidneys from older donors (aged >60 years) that were similar to outcomes of younger-donor kidneys.¹³ A similar strategy yielded good post-transplant outcomes when applied to the allocation of older donors in Spain.¹⁴ Such data provide clinicians with reassurance to use such marginal organs that are otherwise at a high risk of discard in the absence of biopsy results. Importantly, incorporating allocation biopsy information along with clinical characteristics requires standardized biopsy criteria, technique, and reporting to ensure that clinicians understand when biopsy adds independent prognostic information in the context of the clinical history (e.g., understanding age-appropriate elevations in global glomerulosclerosis to avoid inappropriate discard of kidneys from older donors with normal-for-age glomerulosclerosis values >20%).¹⁵

Conclusions and future directions

Given the rapidly increasing demand for kidney transplants globally, improving the utilization of organs from less-than-ideal deceased donors is essential. Allocation biopsies can be helpful in specific circumstances to potentially rescue viable kidneys that are otherwise at a high risk of discard. There is currently little evidence for the wholesale elimination of allocation biopsies given the confounded nature of studies reported (Table 1). Instead, we should focus our efforts on developing an evidence base to inform best practices in biopsy use and interpretation. Such studies, including prospective studies of different clinical thresholds, different methods of tissue sampling, tissue processing (e.g., rapid formalin fixation and paraffin embedding vs. frozen sections), and interpretation (centralized, artificial intelligence–assisted reviews), can help optimize the use of allocation biopsies by maximizing their reliability, reproducibility, and prognostic value while maintaining the focus on improving utilization of kidneys that would otherwise have been discarded because of incomplete or unfavorable clinical information.

Table 1|.

Common criticisms of using deceased donor kidney biopsies during allocation and counterarguments

Criticism	Counterarguments
Allocation biopsy findings are not associated with post-transplant outcomes	• Retrospective studies of the prognostic value of allocation biopsies are prone to selection bias toward a null result. • Studies that focus on standardized allocation biopsies, those read by kidney pathologists, or those that are likely to be both biopsied and transplanted have demonstrated an association between biopsy findings and outcomes. • Histology-informed allocation of kidneys from older donors has been demonstrated to achieve similar outcomes with such kidneys compared with kidneys from younger donors.
Performing allocation biopsies increases the risk of kidney discard	• A subset of transplantable kidneys that are likely to be discarded based on clinical characteristics can be “rescued” by providing other objective information supporting utilization. • Simulation studies have demonstrated improved utilization of marginal organs when favorable histologic findings are presented compared with when no biopsy findings are presented. • Reducing inappropriate biopsies on high-quality organs does not obviate the need for additional organ quality assessments for marginal organs.
Allocation biopsy findings are poorly reproducible	• Factors that contribute to reproducibility limitations (frozen section processing, sampling technique, and interpretation by non-kidney pathologists) are not immutable and can be updated to best practices. • Even gold standard biopsies used routinely in clinical practice have imperfect reliability.
Chronic changes seen on allocation biopsies are often the result of age or known donor comorbidities	• Deceased donors often have unknown medical history or “baseline” laboratory values (e.g., baseline creatinine in donor with acute kidney injury), necessitating a way to obtain additional information in circumstances where the donor’s risk of chronic kidney damage is unclear. • Updated biopsy scoring systems that incorporate donor age and medical history can help clinicians double count biopsy findings that only reflect known donor characteristics.